In a packet-based wireless communications system the system receives packets from an upper layer (for example, from an Internet Protocol (IP) layer). These packets need then be transmitted over a wireless (e.g., cellular mobile) channel, which may offer different and time varying instantaneous data rates to a specific user. Possible reasons for the variations in the data rate may include: different quality of service (QoS) requirements, different instantaneous channel conditions, and also previously sustained data rates. However, an important point to consider is that data packets received from the upper layer(s) should be split (or partitioned or segmented) into smaller data units so as to fit into the instantaneous data capacity on the physical channel. For this to occur, the packets are segmented such that the smallest segment size fits into the lowest resolution provided by the physical layer. An important reason for the presence of the segmentation is that by requiring the segments to have a fixed size, there exists a well-defined interface between the Medium Access Control (MAC) layer and the Physical layer.
However, the inventors have realized that the conventional static segmentation size places limitations and additional requirements on the control signaling for the MAC layer, and may also reduce the scheduling and link adaptation flexibility at the physical layer.